A content delivery network (CDN) implements a distributive nearby service of content by publishing the content to a network edge. In the CDN, content cache servers are arranged on network nodes. A CDN center control system guides a user request to a best service node in real time according to information, such as network traffic, connection and load condition of the nodes and a distance to a user. For the user, using a CDN system shortens response time of a service request and improves stability of data transmission, which thereby improves network service quality.
In the CDN, a client sends a file request to a CDN edge cache server. After receiving the request, the CDN edge cache server determines whether the file is temporarily stored in a local server. If the file is temporarily stored in the local server, the CDN edge cache server directly sends the file to the client; if the file is not temporarily stored in the local server, the CDN edge cache server makes a request to a content source server or an upper-level cache server for obtaining the file, and after obtaining the file, forwards the file to the client.
Currently, a lower-level cache server makes a request to the upper-level cache server or the content source server for obtaining a target file. The upper-level cache server or the content source server sends a file fragment to the lower-level cache server at a preset sending rate of the file fragment, failing to meet diversified downloading requirements of the lower-level cache server, which thereby cannot provide the user with a flexible streaming service.